406 E. HANBURT HANKIN. 



results. On adding the alcoholic Spiller's purple a whitish 

 tinge was produced, as if milk had been added to the solution. 

 This I found to be due to minute globules of aniline. I also 

 found that these globules have a great affinity for the dye, for 

 if aniline or any other oily liquid that can dissolve Spiller's 

 purple is shaken up in a test-tube with a watery solution of 

 this dye nearly all the colouring matter is removed from the 

 solution and dissolved in the oily liquid.^ 



I then tried with watery solutions of Spiller's purple, and 

 obtained successful results, in the following manner : 

 Method B. The materials required are : 



(1) Saturated watery solution of Spiller's purple. 



(2) Saturated alcoholic solution of the same dye. 



It is very important that both these solutions should be 

 saturated. The best way to effect this result is to boil the 

 dye with the solvent used in each case, allow the mixture to 

 cool, and then filter. 



(3) Absolute alcohol, benzine and clove-oil mixture, and 

 benzine as in Method a. 



(4) Eosin dissolved in oil of cloves made by mixing about as 

 much eosin as can be lifted on the point of a penknife with a 

 watch-glass full of oil of cloves. The mixture should be used 

 fresh, as after standing a quantity of the eosin is precipitated 

 in crystals, especially if any acid fumes or traces of picric and 

 other acids are about. 



Modus operandi. — The sections are removed from spirit 

 and placed in the watery Spiller's purple. 



At once an equal bulk of alcoholic Spiller's purple is dropped 

 in from a pipette. The sections are then dehydrated in abso- 

 lute alcohol as quickly as possible and removed to the benzine 

 and clove-oil mixture. 



Sometimes it may be necessary to dehydrate, not in alcohol, 

 but in alcohol saturated with Spiller's purple. By this means 



^ All dyes that stain nuclei also stain the globules of an oil emulsion. 

 Hence when staining — at any rate by these methods — care must be taken that 

 no trace of oil of cloves or any other oil is introduced (on needles, section- 

 lifters, &c.) into the staining solution. Mere traces will often spoil the result. 



